This competitive grant renewal application proposes novel experiments to test a model of depression induced emotional dysregulation and differential changes with medication versus cognitive behavioral therapy (CBT) treatment. This project proposes: 1) Depression-related dysfunction in both implicit and explicit regulation of negative affect is associated with hyperactivations in limbic structures that in turn produce impairment in prefrontal cortex activation. This is evidenced in abnormalities in activations in amygdala and DLPFC and a loss of the normal anticorrelation between limbic and control structures; 2) During regulation of negative affect regions in dorsal cingulate and middle frontal gyrus show hyperactivity, perhaps resulting from inefficient activation. 3) Correlation in activity among limbic structures decreases; and 4) This loss of correlation can be ameliorated by the repeated practice of reframing negative emotional material during CBT. At baseline and following treatment with CBT (n=30) or antidepressants (n=30) we will evaluate regional brain function in depressed subjects (n=60) compared with controls (n=30) in two tasks and a resting state fMRI acquisition. Depressed subjects will be randomized to treatment with either a course of antidepressant treatment with escitalopram or a 12-session course of CBT, and a second fMRI session will be conducted. CBT and antidepressant treatment will be compared with respect to changes in regional activations and correlations among activity in brain regions. 1) We hypothesize that both CBT and antidepressants will decrease amygdala activity and increase DLPFC activity following treatment in both the Conflict and Regulate tasks. We will correlate regional activations during both tasks. 2) We hypothesize that similar to antidepressants, CBT will reinstate the normal anticorrelation between DLPFC and amygdala in both task-based and resting state correlations. We hypothesize that unlike antidepressants, CBT will also restore normal correlations among limbic structures. We will correlate individual scores on key baseline variables (including HAMD scores, MASQ scores and BIS/BAS scores) with amygdala activity. 3) We hypothesize that high scores on these variables will be positively correlated with amygdala activity. We further hypothesize that these variables will correlate with amygdala activity following treatment. Significance: Understanding regulation of emotion is crucial to depression research, both for understanding day to day mood regulation and also as the mechanism through which CBT may operate. Our overall goal is to understand the regional contributions to inability to modulate implicit and explicit negative emotions in depression and what the implications are for understanding treatment. The comparison of antidepressant medication with CBT will add important comparative information about how these treatments affect emotional circuitry and will test an explicit model of emotional circuit dysfunction. PUBLIC HEALTH RELEVANCE In this project we will conduct functional magnetic resonance imaging studies in 60 depressed patients and 30 matched comparison subjects. We will examine changes in brain regions following cognitive behavioral therapy (CBT) or antidepressant treatment during three different conditions. One is an emotionally negative conflict task. The second is a task in which people regulate their emotional reaction to negative pictures. The third is during the resting state. Brain activity will be compared before and after treatment for these tasks in both CBT and antidepressant medication. This will add to our understanding of treatment mechanisms for major depression.